I have had this on my mind for awhile. I really believe that a native productivity suite developed as a replacement of Gobe Productive 2.0 seems like a worthy endeavor. Are there any app developers out there that agree? Please only respond if you’re interested in discussing ideas for such a project.
1 Like
I think it would be nice to have a office suite which doesn’t pull in a ton of dependencies. I counted 21 KDE packages to install LibreOffice, plus the many other packages, some of which I didn’t recognise. I’m not saying the wheel should be reinvented, but porting the kitchen sink and everything else from Linux to work on Haiku might not be the best.
3 Likes
I think Gobe’s approach to a productivity suite was unique in some respects. The one killer feature was being able to manage and save a document, spreadsheet, presentation, etc. into a single file making it very tightly integrated. It was also very easy to use (I actually own a copy of the software I bought bundled with BeOS 5 back in 2000). The great part is, we have a good blueprint to start with and that’s the original version itself, which still runs on Haiku to some extent.
It’s a very appealing idea, but I suspect that either you don’t appreciate how much work will be involved to bring it in line with competing products, or you would be satisfied with something rather less capable than, for instance, LibreOffice.
And if it’s the latter, I wonder how many people would want to use it.
Also it make use of BeOS/Haiku translation kit… But used in translate Word, excel and so on.
1 Like
I have had this idea as well, of doing a greenfield clone of Gobe Productive. I find the prospect of implementing the various components of an office suite interesting as well, though to be honest I find a lot of things interesting 
Limiting the scope to the features Gobe had would make it more realistic. It is probably still a lot of work. But maybe a lot of fun too.
Another thing that would add a bit of complexity but would be innovative and help something like this compete with Google Docs and similar is incorporating some ideas of Local-First Software, which to summarize just means the software puts all the data locally as a first class thing (like normal software before the web), but it also has sharing capabilities built in. This is still a new area and most of the prototypes seem to be in JavaScript. But this could likely be an add-on thing and I would not worry about it until some of the basics are done.
Also I’m now too a point where I don’t love the idea of making a big piece of software like this in C++, so I would seriously consider Rust or Zig or similar, but of course there would need to be bindings to the BeOS/Haiku APIs, and I would want to make it first class on Haiku. But maybe much of the engine and “business logic” could be in a more modern language. This also might get people interested outside of the community. I realize this may be somewhat controversial, but we do have a Rust port and when I mentioned Zig I got a lot of good feedback. Having a big app like this done with one of those languages would really prove their viability as a Haiku application language (or maybe it turns out they don’t work so well.)
Lastly with all this said: I tend to have a lot of ideas, and a lot of things I want to do related to Haiku, but not a lot of time or track record of actually doing things. Some of this is bad time management but a lot of it is just the usual job/family obligations. But I still think this is worth planning out and maybe if I find the time I could try out some prototypes for various pieces of it.
3 Likes
Google Docs is pretty limited but a lot of people use it. Also many people are starting to tire of slow, bloated, over-complicated software. It might be really viable if it was something lightweight which can satisfy the Pareto (80/20) principle with the most important features and maybe also have the sharing aspects of Google Docs using that Local-First Software idea I mentioned.
If it was Haiku only and really well done it would provide another reason to use Haiku over other systems.
People could still fall back to OpenOffice if they needed something more complicated.
2 Likes
I have over the years known many office suites and spreadsheet programs, and I have always wanted something more powerful. I might well be happy to use a text editor to write notes, but I use spreadsheets for work and nothing less than Excel/LibreOffice will suffice.
There has been a lot of discussion an thinking about this (dont remeber the thread).
To develop a own Wordprozessor as a supercomplexe task wich will last verry long (just look at apiword)
There has been a “idea” collection back in the days…
[Update] Args that was not the haiku word processor ideas collection just the template lol
and the plan was to start a “Poetry” office
But i would rate this projekt verry unrealistic.
i would not agree here. I needed to switch from Excel to Google docs beacuse google docs provided much more powerfull posibility to process data … e.g. importrange works for years in google docs while in office365 also after days of reserache there where no suitable solution.
I am very surprised. Can you give more details?
Of course that several people would love to have a “Gobe-like” small and native office suite. But sincerely, that could be a really, really big target.
If someone would like to do something like that, my suggestion is to take some work that is already done, like SumIt spreadsheet, and start to improve that. Spreadsheets also have a more standarized way to share information (.CSV files), so creating something that could be work together with other more professional products could be easier.
And about Word processing: trying to get something that could save complex Word documents, could be a without ending moving target (actually even LibreOffice have some problems dealing with MS Office compatibilty). Instead of that, getting something like an improved version of StyleEdit, with image support, and posibility to export to RTF and simple HTML documents, could be enough.
1 Like
We have several Spreadsheet files which need to be updated when you enter a cell so if you input Value A into sheet ABC it need to show up in Sheet DEF to calculate Value B 
Was this confusing enought? 
There was brainstorming to use replicants. So a Sheet of Paper wich accepts SumIt as a Replicant, a Word processor as Replicant (like improoved StyleEdit) and something like ShowImage
Have you tried using Access? This looks more like a database problem than an Excel problem.
I’d be curious of what the Gobe developers had to build custom that didn’t already have BeOS API support out of the box and see if there is some opportunity to fill those gaps. Also, it would be great to do a lot of prototyping of the various UI/UX components that could become libraries for use in development in other applications. I guess Gobe Productive is still inline with Haiku’s HIG or maybe some subtle improvements can be made to be more inline making it an ambassador app for Haiku.
3 Likes
I think the BFS query system makes it almost an integrated database already. Why not just build on that?
The Medo Haiku native app shows that these days its not impossible for a single developer to build quite complex apps. For something like Gobe Productive, (or a web browser or any other app), you’d require a crowd funding initiative to finance a full time developer. The app can remain free / open source, but does require paid development since its not the typical scratch my itch project.
Disclaimer - Medo was a scratch my itch project for me personally. There are other Haiku native projects I have in various states of completeness (including an actor based browser using C++17 design, which can already display pages from scratch, but will require 2 years to be usable). But there is currently a more interesting itch I’m scratching. Paid development changes focus.
8 Likes
If this is going to be an exercise in Zig or Rust programming, I could use the practice. Otherwise it will just be another C++ debugging nightmare. I’ve never used GoBe Productive so just making a native productivity suite of simple origins sounds ok.
Along those lines I can think of some reusable coding techniques based on the C++ style of designs of inheritance but hopefully in Rust because there are C++ to Rust converters that just use a lot of unsafe directives in the code while Zig assumes a C API and is designed to be used as a cross-compiler.
If we can’t agree on that much, I’d better stick to my existing project: a Yab to C++ transpiler.
2 Likes
Maybe it’s just me… but an office thingamabob really isn’t what Haiku is missing anymore. Yeah LibreOffice and Calligra need a lot of dependencies, but they’re the closest we get to Microsoft’s Office and to this day I’m still like wow 
that someone legit ported LO6 over! That’s crazy, I know that took a lot of work to do, I’m really grateful to who did it
And I guess what I’m saying is sure, an app dev could build a new set of apps like for writing, presenting, sheets to solve the dependencies problem but idk how that’d top what LO and Calligra do already. It feels like that Lego has popped into place and there’s more missing ones to put in 
imho what Haiku needs is improvements to how it connects with the world. There was someone who’d found a way to get the webcam working over a connection — that’s needed especially after the pandemic when everyone wanted to connect with teachers, coworkers, and so on.
Beyond this, and where I’m going with this post, how many modern apps (like from the stores) can Haiku run? The Depot’s great, but I propose that Haiku should also have a spot with web apps to solve this problem; it’s possible to login to Spotify, Instagram, YouTube, Google Docs, etc through a browser on several platforms already, it’s not crazy to think something like this couldn’t work on Haiku. And if another idea I’m thinking of actually works, this might be something I work on as well. Mainly bc I’m more the apps sort, so I can’t contribute anything to the system… but there might be other things I can start working on to help out